Exploring the conformational landscape of adenylate kinase and beyond with protein folding models

Protein folding models have revolutionized structure prediction but struggle to capture conformational flexibility. Recent studies perturb inputs or parameters to sample alternative conformations, while diffusion-based approaches generate conformational ensembles directly. While individual generative models have been benchmarked against molecular dynamics (MD) data, a systematic comparison across diverse methodologies remains scarce, and validation of sub-domain dynamics is still limited. Here, we present a systematic benchmark of nine methods across 20 monomeric proteins with active and inactive states. We extend the pairwise aligned error metric to ensembles and reveal that protein identity exerts a non-negligible influence on model performance. Focusing on Adenylate Kinase, a well-studied enzyme with extensive MD data, we find that Chai-1 performs the best in recovering known conformations, identifying mobile regions, and capturing plausible intermediate conformations. These results highlight the potential of generative models as efficient alternatives to MD for exploring protein conformational dynamics and provide a rigorous benchmark for sampling the protein conformational landscape.